[M3devel] Yet Another OS in Modula-3, AVOCA, in x-kernel

Daniel Alejandro Benavides D. dabenavidesd at yahoo.es
Wed Dec 7 21:20:46 CET 2011


Hi:
yes, uhm I think I see your point is, whether we would be using a Modula-3 OS or not in daily use, and as it happens there was a project on that for ARM based workstation, but didn't panned out, due slow down in performance in HW intrinsics, and some Company's budgeting concerns made it not to be  completed.
It was supposed to be a 2k Acorn Archimedes:
http://neil.franklin.ch/Usenet/alt.folklore.computers/20010824_Not_A_RISC_By_Thursday

http://groups.google.com/group/comp.sys.acorn/browse_thread/thread/6bc26531c038159f/b387b6b1c9abbfd7

This budget is the same others companies tried to achieve with Gnu/Linux, and GNU/Hurd was supposed to be a modern of the day OS, wih Mach based microkernel:
http://ieeexplore.ieee.org/Xplore/login.jsp?url=http%3A%2F%2Fieeexplore.ieee.org%2Fiel1%2F2225%2F1156%2F00028082.pdf%3Farnumber%3D28082&authDecision=-203

Even you could get one working with a TUBE co-processor of the BBC Micro I believe, so you could put a very cheap system with enough productivity to be a workstation:
http://www.osnews.com/thread?189153

DEC tried sending RR and VR via mail service, but you ca judge whether or not they achieved their goal. IMHO it did vastly, since you couldn't get such a good system in the market place of the day

BTW, many people are putting their efforts on reviving the BBC in some time ago.

Thanks in advance

--- El lun, 5/12/11, felipe valdez <felipevaldez at gmail.com> escribió:

De: felipe valdez <felipevaldez at gmail.com>
Asunto: Re: [M3devel] Yet Another OS in Modula-3, AVOCA, in x-kernel
Para: "Daniel Alejandro Benavides D." <dabenavidesd at yahoo.es>
Fecha: lunes, 5 de diciembre, 2011 18:33

READ: I WANT STUFF FOR FREE.GIVE ME STUFF

On Mon, Dec 5, 2011 at 5:41 PM, Daniel Alejandro Benavides D. <dabenavidesd at yahoo.es> wrote:


Hi all:
Given the existence of an Obliq port over CORBA, and the current (perhaps broken) NetObj implementation, and the probable port on Modula-3 x-kernel, then it has sense to actually provide over that number of platforms a Modula-3 to Obliq JIT interpreter, since there already several virtual machines for it and yet two of them (one JIT) with gcc target and based interpreter.


As it happens x-kernel has been ported in the past for OSF/1, Linux and Solaris, I guess if we can host that ports in a cm3/m3-avoca and cm3/m3-sys/m3toobliq then we can given Obliq available packages have an CM3-IDE Obliq installation part to execute code everywhere not just in local machine, with graphics, etc.


If so a question would be Elego willing to link DEC-SRC research reports and technical notes on the Obliq stuff (SRC-122., SRC-128A, SRC-129, SRC-130A, SRC-138,
 SRC-141A, SRC-142, SRC-48 and SRC-TN-1997-029)
Let me know your opinions for such a thing.

Thanks in advance

--- El sáb, 3/12/11, Daniel Alejandro Benavides D. <dabenavidesd at yahoo.es> escribió:



De: Daniel Alejandro Benavides D. <dabenavidesd at yahoo.es>


Asunto: Re: [M3devel] Yet Another OS in Modula-3, AVOCA, in x-kernel
Para: felipevaldez at gmail.com, "Jay K" <jay.krell at cornell.edu>


CC: "m3devel" <m3devel at elegosoft.com>
Fecha: sábado, 3 de diciembre, 2011 20:50



Hi all:
Dr. Nelson (of course
 by the time of his thesis was the pinnacle of AI developments) was and is very ahead of its time (still you can argue that).
Of course ESC/Modula-3 and specifically data abstraction and information hiding verification is so much ahead of our times, that is anything but using the available technology is as many years as DEC-SRC collaborators on ESC programmers plus a number of years of that had not been available.


By that time say in many years we would have a complete system if we were enough smart to scale it to network object verification technology, etc. And consider x-kernel influenced by DEC-SRC Taos, etc, so if you add both sums, I happen to think that is is already ahead of its time. Also consider, several DEC-SRC members are Turing awards and who know whom more by the time we finish it.



Thanks in
 advance

--- El sáb, 3/12/11, Jay K <jay.krell at cornell.edu> escribió:



De: Jay K <jay.krell at cornell.edu>
Asunto: RE: [M3devel] Yet Another OS in Modula-3, AVOCA, in x-kernel
Para: dabenavidesd at yahoo.es, felipevaldez at gmail.com


CC: "m3devel" <m3devel at elegosoft.com>
Fecha: sábado, 3 de diciembre, 2011 20:03




I agree based on the year, it is probably not very valuable.
It is also, I'm sorry, "just" a Ph. D. thesis. It might be a big complete working system.
It might be a barely working prototype, just enough to suggest some lessons learned.


 
 
Either way, please remember, that some of the critical parts of an operating
system are drivers, and their surrounding framework.
There are some less-driver related aspects, like a file system, a networking stack, a scheduler.


It is a lot of work. A lot.
 
I have to run.
 - Jay 
Date: Sun, 4 Dec 2011 00:51:31 +0000
From: dabenavidesd at yahoo.es
To: felipevaldez at gmail.com


CC: m3devel at elegosoft.com
Subject: Re: [M3devel] Yet Another OS in Modula-3, AVOCA, in x-kernel



Hi all:
How can you say a 20 years time is not a small gap, did you consider that by this time we had been on 20 years of Software Crisis (as we happen to have today, and yet GNU/Linux was created by that time), yet this guys new we had something to do in Optical Fiber high speed networks, and also 10 years of artificial intelligence lack of invention by the time (as we had 30 years a couple of years ago of productivity-less research).


So basically this time is reduced for me if you consider the previous times also a little, but anyway, I give you a better answer if one doesn't know who used it? What is that suppose to mean, that nobody used it?


The same is true for the 30 of lack of artificial intelligence and 40 years of software crisis, if we
 don't get things done quickly will be arriving at a software  "fiasco", where everything on it is
 usable, and system perform quickly than never but can't get systems to work in a efficient fashion (CPU overheating).

Thanks in advance 

--- El sáb, 3/12/11, felipe valdez <felipevaldez at gmail.com> escribió:



De: felipe valdez <felipevaldez at gmail.com>
Asunto: Re: [M3devel] Yet Another OS in Modula-3, AVOCA, in x-kernel


Para: "Daniel Alejandro Benavides D." <dabenavidesd at yahoo.es>
Fecha: sábado, 3 de diciembre, 2011 18:46





On Sat, Dec 3, 2011 at 4:58 PM, Daniel Alejandro Benavides D. <dabenavidesd at yahoo.es> wrote:


Hi all:

whether we want it or not, perhaps the best opportunity we have is already a complete operating operating system developed in Modula-3 for developing network applications in http://arizona.openrepository.com/arizona/bitstream/10150/185158/1/azu_td_9103022_sip1_m.pdf









whether you like it or not that OS is probably incompatible with everything out there, and is probably unsuitable for production work.but perhap's I'm just being mean for the sake of being mean




let's have a look at it.


...

oh, goodie!an article 20 years old.or, 200, in internet years.




I've lost interest already.
after further examination, I can see it is not only ancient, is seems to be very interesting...
however, don't you have anything, more recent?,



as in, not previous-century?








 for Linux, OSF, etc, under the x-kernel.



Can we create a branch to develop operating system engine line (you know we don't need any more platforms if we happen to have what is really on any network environment a browser and a file system service), for developing GUIs, things and everybody could license it in whatever license whoever wants to use it might be used safely in any environment, you don't like DEC licenses appearance OK, but for me but if anyone likes them then he can develop in a system that still runs code in a DEC environment and then since everything are objects moving around like process one can define them absolutely license-independent.







I think OS developers in everywhere would love it (I remember an OOP class, where the class Prof. was explaining whether creating an OS  would be called a good class final project, and he seemed to think it wasn't, based on a former classmate of him thought that it was).







If we could create an enough smart environment then one could think in that again. I know several browsers including one outside of DEC that would be appealing to accomplish the work (yes a complete browser engine system in Modula-3).







And the project will be hosting the Obliq based tools in that browser for others as well, but mainly educational/research services for anyone in the net with a bright idea for a system application or just want to experiment with a Modula-3 Obliq super-optimizer at source level or whatever want to do it separately as well:





http://www.cs.utexas.edu/ftp/techreports/tr92-27.pdf



See for example see for M-F4 and G-V8:

http://bellard.org/jslinux



Thanks in advance








-- 
312-444-2124Skype: f3l.headhunterCasa: 8043901





 		 	   		  



-- 
312-444-2124Skype: f3l.headhunterCasa: 8043901






-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://m3lists.elegosoft.com/pipermail/m3devel/attachments/20111207/f32b1125/attachment-0002.html>


More information about the M3devel mailing list